Grok by xAI
Grok chat, Responses and Messages calls, embeddings, image generation, tokenization, models and files via the xAI API.
About
Grok by xAI ships in the w6w first-party pack. It declares 14 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.xai- Version
- 0.1.2
- Author
- w6w
- Licence
- MIT
- Categories
- AI & Machine Learning
Overview
Grok by xAI puts xAI’s Grok models inside a workflow. Chat completions, the stateful Responses API and the Anthropic-compatible Messages endpoint are each a step, so a workflow can summarize, classify, extract or draft with whichever request shape it already speaks. A stored Responses reply can be fetched or deleted later by id, and the Responses API can continue a conversation from a previous response id without resending the history.
Embeddings turn text into vectors for search and clustering, image generation produces images from a prompt (aspect ratio and resolution apply to the grok-imagine models), and token counting lets a workflow size a prompt before sending it. The model catalog can be listed — including the richer language-models listing with modalities and aliases — so a workflow’s model choice never has to be hardcoded. Files already uploaded to xAI storage can be listed, inspected and deleted.
A fit for adding a Grok step to a backend automation, or for routing the same prompt to xAI alongside other LLM vendors.
Build with Grok by xAI
Three routes to the same 14 actions. The Workflow tab is generated from Grok by xAI's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
Create Message (Anthropic-compatible)
create-message Create a message through xAI's Anthropic-compatible endpoint (POST /v1/messages).
List Language Models
list-language-models List chat and image-understanding models with modalities and aliases (GET /v1/language-models).
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Grok by xAI's real ids.
{
"manifestVersion": "2",
"name": "xai-example",
"steps": [
{
"id": "chat-complete",
"uses": {
"app": "io.w6w.xai",
"action": "chat-complete",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"model": "<model>"
}
}
]
}Here are some of the things you can do
-
Chat Completion
performchat-complete -
Create Embedding
performcreate-embedding -
Create Message (Anthropic-compatible)
performcreate-message -
Create Response
performcreate-response -
Generate Image
performgenerate-image
+9 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Grok by xAI, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Grok by xAI. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "chat-complete",
payload: {
model: "<value>",
// prompt: "<value>",
// system: "<value>",
// messages: "<value>",
// temperature: "<value>",
// topP: "<value>",
// maxCompletionTokens: "<value>",
// reasoningEffort: "<value>",
// seed: "<value>",
// responseFormat: "<value>",
// tools: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action chat-complete --payload '{"model":"<value>"}'Give an AI agent Grok by xAI — without giving it Grok by xAI's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.xai#chat-complete",
"input": {
"model": "<model>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Grok by xAI connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
A tool surface scoped to the caller
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
A durable workflow in one call
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Health-aware discovery
Grok by xAI's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Grok by xAI itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
Grok by xAI declares its own checks, so its health is a property of the app rather than something the host guesses at.
xAI API status
Reads status.x.ai's RSS incident feed for open incidents on the API components.